[html5] r2084 - [] (0) Further work on the event loop front. (<script>, database API, remote eve [...]
whatwg at whatwg.org
whatwg at whatwg.org
Tue Aug 19 02:00:42 PDT 2008
Author: ianh
Date: 2008-08-19 02:00:41 -0700 (Tue, 19 Aug 2008)
New Revision: 2084
Modified:
index
source
Log:
[] (0) Further work on the event loop front. (<script>, database API, remote events, and other bits and bobs)
Modified: index
===================================================================
--- index 2008-08-19 08:21:09 UTC (rev 2083)
+++ index 2008-08-19 09:00:41 UTC (rev 2084)
@@ -18825,10 +18825,10 @@
href="#task-queues" title="task queue">task queues</a>, the <dfn
id=media11>media element event task source</dfn> for asynchronous events
and callbacks, and the <dfn id=media12>media element new resource task
- source</dfn> for handling implicit loads. Unless otherwise specified, all
- the tasks <a href="#queue" title="queue a task">queued</a> in this section
- and its subsections use the <a href="#media11">media element event task
- source</a>.
+ source</dfn> for handling implicit loads. Unless otherwise specified, the
+ <a href="#task-source">task source</a> for all the tasks <a href="#queue"
+ title="queue a task">queued</a> in this section and its subsections is the
+ <a href="#media11">media element event task source</a>.
<h5 id=error><span class=secno>4.7.10.1. </span>Error codes</h5>
@@ -26560,9 +26560,9 @@
href="#the-javascript" title="javascript protocol"><code
title="">javascript:</code> URL</a>, then the user agent must not,
despite the requirements in the definition of the <a href="#fetch"
- title=fetch>fetching</a> algorithm, actually execute the given script,
- and instead the user agent must act as if it had received an empty HTTP
- 400 response.</p>
+ title=fetch>fetching</a> algorithm, actually execute the given script;
+ instead the user agent must act as if it had received an empty HTTP 400
+ response.</p>
<p>Once the fetching process has completed, and the script has <dfn
id=completed>completed loading</dfn>, the user agent will have to
@@ -26642,9 +26642,9 @@
<p><dfn id=when-a>When a script completes loading</dfn>: If the script's
element was added to one of the lists mentioned above and the document is
still being parsed, then the parser handles it. Otherwise, the UA must run
- the following steps as soon as as any other scripts that may be executing
- have finished executing:</p>
- <!-- XXX queue -->
+ the following steps as the <a href="#tasks" title=concept-task>task</a>
+ that the <a href="#networking">networking task source</a> places on the <a
+ href="#task-queues">task queue</a>:
<dl class=switch>
<dt>If the script's element was added to the <dfn id=list-of>list of
@@ -30850,6 +30850,11 @@
<h4 id=processing1><span class=secno>4.12.6 </span>Processing model</h4>
<!-- ua side -->
+ <p>The <span>event source</span> for all <a href="#tasks"
+ title=concept-task>tasks</a> <a href="#queue" title="queue a
+ task">queued</a> by algorithms in this section and its subsections is the
+ <dfn id=template1>template event source</dfn>.
+
<h5 id=the-originalcontent><span class=secno>4.12.6.1. </span>The <code
title=dom-originalContent><a
href="#originalcontent">originalContent</a></code> DOM attribute</h5>
@@ -30895,10 +30900,8 @@
steps below.
<p>Once the DOM in question has been parsed, assuming that it indeed can be
- parsed and does so successfully, the user agent must wait for no scripts
- to be executing, and as soon as that opportunity
- arises<!-- XXX queue; xref to a strict definition of this -->, run the
- following algorithm:
+ parsed and does so successfully, the user agent must <a
+ href="#queue">queue a task</a> to run the following algorithm:
<ol><!-- precondition: the originalContent is null and the element
contains its original content. -->
@@ -31021,10 +31024,8 @@
below.
<p>Once the DOM in question has been parsed, assuming that it indeed can be
- parsed and does so successfully, the user agent must wait for no scripts
- to be executing, and as soon as that opportunity
- arises<!-- XXX; queue xref to a strict definition of this -->, run the
- following algorithm:
+ parsed and does so successfully, the user agent must <a
+ href="#queue">queue a task</a> to run the following algorithm:
<ol>
<li>
@@ -31128,21 +31129,21 @@
<p>An element with a non-null <code title=dom-templateElement><a
href="#templateelement">templateElement</a></code> is also said to be a
- <dfn id=template1>template tree user</dfn> of the node identified by the
+ <dfn id=template2>template tree user</dfn> of the node identified by the
element's <code title=dom-templateElement><a
href="#templateelement">templateElement</a></code> attribute, as well as
all of that node's children.
- <p>Nodes that have one or more <a href="#template1" title="template tree
+ <p>Nodes that have one or more <a href="#template2" title="template tree
user">template tree users</a> associated with them (as per the previous
- paragraph) are themselves termed <dfn id=template2 title="template tree
+ paragraph) are themselves termed <dfn id=template3 title="template tree
component node">template tree component nodes</dfn>.
- <p>Whenever a <a href="#template2">template tree component node</a> changes
+ <p>Whenever a <a href="#template3">template tree component node</a> changes
its name or value, or has one of its attributes change name or value, or
has an attribute added or removed, or has a child added or removed, the
user agent must <span>update the generated content</span> of all of that
- node's <a href="#template1" title="template tree user">template tree
+ node's <a href="#template2" title="template tree user">template tree
users</a>.
<p class=note>In other words, user agents update the content generated from
@@ -37632,19 +37633,16 @@
href="#storage0">Storage</a></code> object never changes while a script is
executing, other than in a way that is predictable by the script itself.
- <p>There are various ways of implementing this requirement. One is that if
- a script running in one browsing context accesses a local storage area,
- the UA blocks scripts in other browsing contexts when they try to access
- the local storage area for the <a href="#same-origin">same origin</a>
- until the first script has executed to completion <!--
- xxx queue -->.
- (Similarly, when a script in one browsing context accesses its session
- storage area, any scripts that have the same top level browsing context
- and the <a href="#same-origin">same origin</a> would block when accessing
- their session storage area until the first script has executed to
- completion.) Another (potentially more efficient but probably more
- complex) implementation strategy is to use optimistic transactional script
- execution. This specification does not require any particular
+ <p class=note>There are various ways of implementing this requirement. One
+ is to just have one <a href="#event3">event loop</a> for all <a
+ href="#browsing1" title="browsing context">browsing contexts</a>. Another
+ is that if a script running in one browsing context accesses a storage
+ area, the user agent blocks scripts in other browsing contexts when they
+ try to access the same storage area until the <a href="#event3">event
+ loop</a> running the first script has completed running the task that
+ started that script. Another (potentially more efficient but certainly
+ more complex) implementation strategy is to use optimistic transactional
+ script execution. This specification does not require any particular
implementation strategy, so long as the requirement above is met.
<h4 id=sql><span class=secno>5.10.2 </span>Database storage</h4>
@@ -38087,9 +38085,10 @@
href="#changeversion">changeVersion()</a></code> method.)
<li>
- <p><!-- XXX queue --> Invoke the <i>transaction callback</i> with the
- aforementioned <code><a href="#sqltransaction">SQLTransaction</a></code>
- object as its only argument.
+ <p><a href="#queue">Queue a task</a> to invoke the <i>transaction
+ callback</i> with the aforementioned <code><a
+ href="#sqltransaction">SQLTransaction</a></code> object as its only
+ argument, and wait for that task to be run.
<li>
<p>If the callback couldn't be called (e.g. it was null), or if the
@@ -38124,12 +38123,12 @@
object that represents the result of the statement.
<li>
- <p><!-- XXX queue --> If the statement has a result set callback,
- invoke it with the <code><a
+ <p>If the statement has a result set callback, <a href="#queue">queue a
+ task</a> to invoke it with the <code><a
href="#sqltransaction">SQLTransaction</a></code> object as its first
argument and the new <code><a
href="#sqlresultset">SQLResultSet</a></code> object as its second
- argument.
+ argument, and wait for that task to be run.
<li>
<p>If the callback was invoked and raised an exception, jump to the
@@ -38145,12 +38144,12 @@
<ol>
<li>
- <p>If the statement had an associated error callback, then invoke that
- error callback with the <code><a
- href="#sqltransaction">SQLTransaction</a></code> object and a newly
- constructed <code><a href="#sqlerror">SQLError</a></code> object that
- represents the error that caused these substeps to be run as the two
- arguments, respectively.
+ <p>If the statement had an associated error callback, then <a
+ href="#queue">queue a task</a> to invoke that error callback with the
+ <code><a href="#sqltransaction">SQLTransaction</a></code> object and a
+ newly constructed <code><a href="#sqlerror">SQLError</a></code> object
+ that represents the error that caused these substeps to be run as the
+ two arguments, respectively, and wait for the task to be run.
<li>
<p>If the error callback returns false, then move on to the next
@@ -38178,16 +38177,18 @@
last step.
<li>
- <p>Invoke the <i>success callback</i>.
+ <p><a href="#queue">Queue a task</a> to invoke the <i>success
+ callback</i>.
<li>
<p>End these steps. The next step is only used when something goes wrong.
<li>
- <p>Call the <i>error callback</i> with a newly constructed <code><a
- href="#sqlerror">SQLError</a></code> object that represents the last
- error to have occurred in this transaction. Rollback the transaction.
- Any still-pending statements in the transaction are discarded.
+ <p><a href="#queue">Queue a task</a> to invoke the <i>error callback</i>
+ with a newly constructed <code><a href="#sqlerror">SQLError</a></code>
+ object that represents the last error to have occurred in this
+ transaction. Rollback the transaction. Any still-pending statements in
+ the transaction are discarded.
</ol>
<h4 id=disk-space><span class=secno>5.10.3 </span>Disk space</h4>
@@ -42927,12 +42928,24 @@
<dfn id=list-of3 title=concept-eventsource-list>list of event
sources</dfn> that are registered for that object.
- <p>When a new URI is added to this list, the user agent should, as soon as
- all currently executing scripts (if any) have finished
- executing<!-- XXX queue -->, and if the specified URL isn't removed from
- the list before they do so, <a href="#fetch">fetch</a> the resource
- identified by that URL.
+ <p>When a new <a href="#absolute">absolute URL</a> is added to this list,
+ the user agent should <a href="#queue">queue a task</a> to run the
+ following steps with the new <a href="#absolute">absolute URL</a>:
+ <ol>
+ <li>
+ <p>If the entry for the new <a href="#absolute">absolute URL</a> has been
+ removed from the list, then abort these steps.
+
+ <li>
+ <p><a href="#fetch">Fetch</a> the resource identified by that <a
+ href="#absolute">absolute URL</a>.</p>
+
+ <p>As data is received, the <a href="#tasks" title=concept-task>tasks</a>
+ queued by the <span>networking event source</span> to handle the data
+ must consist of following the rules given in the following sections.</p>
+ </ol>
+
<p>When an event source is removed from the list of event sources for an
object, if that resource is still being fetched, then the relevant
connection must be closed.
@@ -43239,9 +43252,12 @@
name</var> buffer to the empty string.
<li>
- <p>Dispatch the newly created event at the <code><a
- href="#remoteeventtarget">RemoteEventTarget</a></code> object to which
- the event stream is registered.
+ <p><a href="#queue">Queue a task</a> to dispatch the newly created event
+ at the <code><a href="#remoteeventtarget">RemoteEventTarget</a></code>
+ object to which the event stream is registered. The <a
+ href="#task-source">task source</a> for this <a href="#tasks"
+ title=concept-task>task</a> is the <dfn id=remote>remote event task
+ source</dfn>.
</ol>
<p class=note>If an event doesn't have an "id" field, but an earlier event
Modified: source
===================================================================
--- source 2008-08-19 08:21:09 UTC (rev 2083)
+++ source 2008-08-19 09:00:41 UTC (rev 2084)
@@ -16340,10 +16340,10 @@
title="task queue">task queues</span>, the <dfn>media element event
task source</dfn> for asynchronous events and callbacks, and the
<dfn>media element new resource task source</dfn> for handling
- implicit loads. Unless otherwise specified, all the tasks <span
- title="queue a task">queued</span> in this section and its
- subsections use the <span>media element event task
- source</span>.</p>
+ implicit loads. Unless otherwise specified, the <span>task
+ source</span> for all the tasks <span title="queue a
+ task">queued</span> in this section and its subsections is the
+ <span>media element event task source</span>.</p>
@@ -23935,7 +23935,7 @@
title="javascript protocol"><code title="">javascript:</code>
URL</span>, then the user agent must not, despite the requirements
in the definition of the <span title="fetch">fetching</span>
- algorithm, actually execute the given script, and instead the user
+ algorithm, actually execute the given script; instead the user
agent must act as if it had received an empty HTTP 400
response.</p>
@@ -24035,9 +24035,9 @@
<p><dfn>When a script completes loading</dfn>: If the script's
element was added to one of the lists mentioned above and the
document is still being parsed, then the parser handles
- it. Otherwise, the UA must run the following steps as soon as as any
- other scripts that may be executing have finished executing:</p>
- <!-- XXX queue -->
+ it. Otherwise, the UA must run the following steps as the <span
+ title="concept-task">task</span> that the <span>networking task
+ source</span> places on the <span>task queue</span>:</p>
<dl class="switch">
@@ -28063,6 +28063,12 @@
<h4>Processing model</h4>
<!-- ua side -->
+ <p>The <span>event source</span> for all <span
+ title="concept-task">tasks</span> <span title="queue a
+ task">queued</span> by algorithms in this section and its
+ subsections is the <dfn>template event source</dfn>.</p>
+
+
<h5>The <code title="dom-originalContent">originalContent</code> DOM
attribute</h5>
@@ -28106,10 +28112,8 @@
parse</span> steps below.</p>
<p>Once the DOM in question has been parsed, assuming that it indeed
- can be parsed and does so successfully, the user agent must wait for
- no scripts to be executing, and as soon as that opportunity
- arises<!-- XXX queue; xref to a strict definition of this -->, run the
- following algorithm:</p>
+ can be parsed and does so successfully, the user agent must
+ <span>queue a task</span> to run the following algorithm:</p>
<ol>
<!-- precondition: the originalContent is null and the element
@@ -28235,10 +28239,8 @@
steps below.</p>
<p>Once the DOM in question has been parsed, assuming that it indeed
- can be parsed and does so successfully, the user agent must wait for
- no scripts to be executing, and as soon as that opportunity
- arises<!-- XXX; queue xref to a strict definition of this -->, run
- the following algorithm:</p>
+ can be parsed and does so successfully, the user agent must
+ <span>queue a task</span> to run the following algorithm:</p>
<ol>
@@ -35032,20 +35034,18 @@
executing, other than in a way that is predictable by the script
itself.</p>
- <p>There are various ways of implementing this requirement. One is
- that if a script running in one browsing context accesses a local
- storage area, the UA blocks scripts in other browsing contexts when
- they try to access the local storage area for the <span>same
- origin</span> until the first script has executed to completion <!--
- xxx queue -->. (Similarly, when a script in one browsing context
- accesses its session storage area, any scripts that have the same
- top level browsing context and the <span>same origin</span> would
- block when accessing their session storage area until the first
- script has executed to completion.) Another (potentially more
- efficient but probably more complex) implementation strategy is to
- use optimistic transactional script execution. This specification
- does not require any particular implementation strategy, so long as
- the requirement above is met.</p>
+ <p class="note">There are various ways of implementing this
+ requirement. One is to just have one <span>event loop</span> for all
+ <span title="browsing context">browsing contexts</span>. Another is
+ that if a script running in one browsing context accesses a storage
+ area, the user agent blocks scripts in other browsing contexts when
+ they try to access the same storage area until the <span>event
+ loop</span> running the first script has completed running the task
+ that started that script. Another (potentially more efficient but
+ certainly more complex) implementation strategy is to use optimistic
+ transactional script execution. This specification does not require
+ any particular implementation strategy, so long as the requirement
+ above is met.</p>
@@ -35494,9 +35494,10 @@
title="dom-database-changeVersion">changeVersion()</code>
method.)</p></li>
- <li><p><!-- XXX queue --> Invoke the <i>transaction callback</i>
- with the aforementioned <code>SQLTransaction</code> object as its
- only argument.</p></li>
+ <li><p><span>Queue a task</span> to invoke the <i>transaction
+ callback</i> with the aforementioned <code>SQLTransaction</code>
+ object as its only argument, and wait for that task to be
+ run.</p></li>
<li><p>If the callback couldn't be called (e.g. it was null), or if
the callback was invoked and raised an exception, jump to the last
@@ -35526,10 +35527,11 @@
<li><p>Create a <code>SQLResultSet</code> object that represents
the result of the statement.</p></li>
- <li><p><!-- XXX queue --> If the statement has a result set
- callback, invoke it with the <code>SQLTransaction</code> object
- as its first argument and the new <code>SQLResultSet</code>
- object as its second argument.</p></li>
+ <li><p>If the statement has a result set callback, <span>queue a
+ task</span> to invoke it with the <code>SQLTransaction</code>
+ object as its first argument and the new
+ <code>SQLResultSet</code> object as its second argument, and wait
+ for that task to be run.</p></li>
<li><p>If the callback was invoked and raised an exception, jump
to the last step in the overall steps.</p></li>
@@ -35546,10 +35548,11 @@
<ol>
<li><p>If the statement had an associated error callback, then
- invoke that error callback with the <code>SQLTransaction</code>
- object and a newly constructed <code>SQLError</code> object that
- represents the error that caused these substeps to be run as the
- two arguments, respectively.</p></li>
+ <span>queue a task</span> to invoke that error callback with the
+ <code>SQLTransaction</code> object and a newly constructed
+ <code>SQLError</code> object that represents the error that
+ caused these substeps to be run as the two arguments,
+ respectively, and wait for the task to be run.</p></li>
<li><p>If the error callback returns false, then move on to the
next statement, if any, or onto the next overall step
@@ -35575,15 +35578,17 @@
<li><p>If an error occurred in the committing of the transaction,
jump to the last step.</p></li>
- <li><p>Invoke the <i>success callback</i>.</p></li>
+ <li><p><span>Queue a task</span> to invoke the <i>success
+ callback</i>.</p></li>
<li><p>End these steps. The next step is only used when something
goes wrong.</p></li>
- <li><p>Call the <i>error callback</i> with a newly constructed
- <code>SQLError</code> object that represents the last error to have
- occurred in this transaction. Rollback the transaction. Any
- still-pending statements in the transaction are discarded.</p></li>
+ <li><p><span>Queue a task</span> to invoke the <i>error
+ callback</i> with a newly constructed <code>SQLError</code> object
+ that represents the last error to have occurred in this
+ transaction. Rollback the transaction. Any still-pending statements
+ in the transaction are discarded.</p></li>
</ol>
@@ -40193,12 +40198,29 @@
title="concept-eventsource-list">list of event sources</dfn> that
are registered for that object.</p>
- <p>When a new URI is added to this list, the user agent should, as
- soon as all currently executing scripts (if any) have finished
- executing<!-- XXX queue -->, and if the specified URL isn't removed
- from the list before they do so, <span>fetch</span> the resource
- identified by that URL.</p>
+ <p>When a new <span>absolute URL</span> is added to this list, the
+ user agent should <span>queue a task</span> to run the following
+ steps with the new <span>absolute URL</span>:</p>
+ <ol>
+
+ <li><p>If the entry for the new <span>absolute URL</span> has been
+ removed from the list, then abort these steps.</p></li>
+
+ <li>
+
+ <p><span>Fetch</span> the resource identified by that
+ <span>absolute URL</span>.</p>
+
+ <p>As data is received, the <span
+ title="concept-task">tasks</span> queued by the <span>networking
+ event source</span> to handle the data must consist of following
+ the rules given in the following sections.</p>
+
+ </li>
+
+ </ol>
+
<p>When an event source is removed from the list of event sources
for an object, if that resource is still being fetched, then the
relevant connection must be closed.</p>
@@ -40527,9 +40549,11 @@
<li><p>Set the <var title="">data</var> buffer and the <var
title="">event name</var> buffer to the empty string.</p></li>
- <li><p>Dispatch the newly created event at the
- <code>RemoteEventTarget</code> object to which the event stream is
- registered.</p></li>
+ <li><p><span>Queue a task</span> to dispatch the newly created
+ event at the <code>RemoteEventTarget</code> object to which the
+ event stream is registered. The <span>task source</span> for this
+ <span title="concept-task">task</span> is the <dfn>remote event
+ task source</dfn>.</p></li>
</ol>
More information about the Commit-Watchers
mailing list